Product
I undertook the following product-focused side projects to learn about what it takes to develop an impactful software product from scratch. Below are project descriptions along with links to detailed outlines with my thinking across the problem, solution, development process, final outcome and a post-mortem for each project.
Bloomberg for the masses
Built an investment insights product to bridge the gap between frictionless investing apps such as Robinhood and the often limited investing knowledge of their users.
Debit card point API for financial institutions
Built an API that would give point collecting capabilities to debit cards issued by small to mid-sized financial institutions (e.g. credit unions and regional banks).
Programming
I undertook the following programming-focused side projects to get experience building apps from scratch to solve challenges. This helped me move beyond product planning and syntax to working with classes & objects, refactoring code for speed and efficiency, integrating various libraries, using Git for version control, getting data dashboards up and running online, etc. The projects below are broken up into detailed study projects with write-ups and velocity-driven projects with GitHub and web service links where available (note that they may take a few moments to spool up).
Detailed Study Projects
Gamma exposure calculator
Built a calculator that estimates the gamma exposure of market makers using ingested option chain data. This can be useful for determining if market makers are being forced to buy or sell a certain asset in order to maintain neutral footing.
SPY threshold price calculator based on U.S. Money Stock
Built a calculator based on a regression model to determine how the equity market is trading in relation to the U.S. Money Stock. This can be useful for determining a threshold price for the market, above which it may be overvalued and below which it may be undervalued.
Velocity-driven Projects (GitHub links)
Earnings Call Sentiment Analysis Tool [Demo]
Derives an earnings call sentiment score for given transcript based on the VADER (Valence Aware Dictionary and sEntiment Reasoner) model
Oil Price Predictor LSTM Model
Forecasts crude oil prices based on an LSTM (long short-term memory) RNN (recurrent neural network) model
Japanese Text OCR Computer Vision Tool
Translates Japanese text in an uploaded image to English using the Tesseract OCR (optical character recognition) computer vision engine and the Google Translate api
Los Angeles Metro Live Map [Demo]
Displays a real-time visual representation of train positions across the Los Angeles metro system
Weather Forecast [Demo]
Shows weather forecast for any given location including temperature, precipitation and UV index data
Nikkei Japanese Vocabulary Training [Demo]
Quizzes user on the definitions of the most used vocabulary in the Nikkei newspaper